US 11,991,658 B2
Learning communication systems using channel approximation
Timothy J. O'Shea, Arlington, VA (US); Ben Hilburn, Reston, VA (US); Tamoghna Roy, Arlington, VA (US); and Nathan West, Washington, DC (US)
Assigned to DeepSig Inc., Arlington, VA (US)
Filed by DeepSig Inc., Arlington, VA (US)
Filed on Feb. 17, 2022, as Appl. No. 17/674,020.
Application 16/732,412 is a division of application No. 16/291,936, filed on Mar. 4, 2019, granted, now 10,531,415, issued on Jan. 7, 2020.
Application 17/674,020 is a continuation of application No. 16/732,412, filed on Jan. 2, 2020, granted, now 11,259,260.
Claims priority of provisional application 62/664,306, filed on Apr. 30, 2018.
Claims priority of provisional application 62/637,770, filed on Mar. 2, 2018.
Prior Publication US 2022/0174634 A1, Jun. 2, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. H04W 56/00 (2009.01); G06N 3/044 (2023.01); G06N 3/08 (2023.01); G06N 3/084 (2023.01); G06N 20/00 (2019.01); H04B 17/391 (2015.01); H04L 5/00 (2006.01); H04L 41/14 (2022.01); H04W 16/22 (2009.01); H04W 72/0453 (2023.01)
CPC H04W 56/0035 (2013.01) [G06N 3/044 (2023.01); G06N 3/08 (2013.01); G06N 3/084 (2013.01); G06N 20/00 (2019.01); H04B 17/3912 (2015.01); H04L 5/0005 (2013.01); H04L 41/145 (2013.01); H04W 16/22 (2013.01); H04W 72/0453 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method performed by at least one processor to train at least one machine-learning network to communicate over a communication channel, the method comprising:
transmitting input information through a first communication channel;
obtaining first information as an output of the first communication channel;
transmitting the input information through a second communication channel implementing a channel machine-learning network, the second communication channel representing a model of the first communication channel;
obtaining second information as an output of the second communication channel;
providing the first information or the second information to a discriminator machine-learning network as an input;
obtaining an output of the discriminator machine-learning network;
updating the channel machine-learning network using the output of the discriminator machine-learning network; and
using the second communication channel implementing the updated channel machine-learning network to determine one or more performance metrics that represent an estimate of the performance of the first communication channel.